Paperless operation data transmission method, system and device
By using deep learning algorithms and grid zero-crossing window technology, the problem of noise interference in power distribution lines at new energy power plants has been solved, achieving efficient and reliable paperless data transmission and meeting low latency requirements.
Patent Information
- Application Number
- CN202610029677.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-10
- Publication Date
- 2026-02-06
AI Technical Summary
The power distribution lines of new energy power plants have low signal-to-noise ratios due to noise interference and load fluctuations. Traditional adaptive filtering algorithms have slow convergence speeds and cannot meet the data transmission requirements of low-latency paperless operations.
Noise separation is performed using a convolutional recurrent network model based on deep learning algorithms to generate a clean signal. The simulated waveform is then transmitted during the low-noise window period of the power grid's zero-crossing point. Combined with high-order QAM modulation and spread spectrum modulation, the signal is coupled to the power distribution line through an impedance matching transformer for transmission. The receiving end directly calls the trained model to eliminate interference.
It improves the integrity and reliability of data transmission, reduces processing latency, meets real-time requirements, avoids instantaneous interference caused by load fluctuations, and realizes low-latency paperless data transmission.
Smart Images
Figure CN121485718A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power distribution line transmission technology, and in particular to a paperless operation data transmission method, system and device. Background Technology
[0002] With the advancement of intelligent construction in new energy power plants, paperless operation systems are gradually becoming widespread in wind power, photovoltaic, and other power plants. Mobile devices need to transmit business information such as inspection records, work order instructions, and equipment maintenance data to the management platform in real time. Traditional solutions rely on dedicated communication networks, such as fiber optics or 4G / 5G wireless networks, but these are costly to deploy in remote power plants and are easily constrained by factors such as terrain and weather.
[0003] Power distribution lines, due to their wide coverage and lack of need for additional wiring, have become a viable alternative transmission medium. However, power distribution lines in new energy power plants have the following inherent drawbacks: The high-frequency narrowband noise generated by equipment such as frequency converters and inverters in the station results in a relatively low signal-to-noise ratio. Load fluctuations cause impedance mismatch, resulting in large fluctuations in signal attenuation. Traditional adaptive filtering algorithms have slow convergence speed and cannot meet the needs of low-latency services. Summary of the Invention
[0004] This invention provides a paperless data transmission method, system, and apparatus, which can effectively solve the problems pointed out in the background art.
[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: Paperless data transmission methods utilize power distribution lines, including: Noise separation is performed on the raw electrical signals collected in the power distribution line transmission link to generate a cleaned signal. The noise separation is achieved through a network model trained based on a deep learning algorithm. Based on the environment of the purification signal, the paperless operation data of the mobile terminal device is converted into a simulated waveform suitable for power distribution line transmission; The simulated waveform is coupled to the power distribution line and transmitted during the low-noise window period of the power grid zero-crossing point; The terminal equipment of the power distribution line transmission link reuses the network model to perform noise separation on the received mixed signal in order to eliminate interference; The paperless operation data is restored by decoding the mixed signal after interference elimination.
[0006] Furthermore, the network model is a convolutional recurrent network model, and the noise separation of the original electrical signal includes: The original electrical signal is input into the convolutional layer to extract the time-frequency domain features of the noise and generate a time-frequency feature tensor. The time-frequency feature tensor is input into the recursive layer to capture the time dependency of the noise components and generate a noise dynamic feature sequence. The noise dynamic feature sequence is decoded by a fully connected layer, and the reconstructed interference waveform is output. The purified signal is generated by directly subtracting the interference waveform from the original electrical signal.
[0007] Furthermore, based on the environment of the purified signal, the paperless operation data of the mobile device is converted into an analog waveform suitable for power distribution line transmission, including: The paperless operation data is fragmented and encapsulated into transmission protocol units containing at least a frame header, a checksum, and a sequence number identifier to generate standard data blocks; The purification signal is analyzed to identify frequency bands with a signal-to-noise ratio higher than the threshold, strong interference frequency bands of frequency converters or inverters, and high-risk noise frequency bands. High-order QAM modulation is used in the frequency band where the signal-to-noise ratio is higher than the threshold, and spread spectrum modulation is switched in the strong interference frequency band of the frequency converter or inverter to convert the standard data block into a modulated waveform. The modulated waveform is converted into an analog baseband waveform synchronized with the power frequency cycle; The frequency components corresponding to the high-risk noise bands are filtered out from the simulated baseband waveform, and the energy distribution during the low-noise window period of the power grid zero crossing is enhanced to output the optimized simulated waveform.
[0008] Furthermore, the execution cycle of the signal analysis is 0.5-0.8 seconds, and real-time re-analysis is triggered when a start-up or shutdown event of the frequency converter or inverter is detected.
[0009] Furthermore, the transmission protocol unit also includes a data length identifier; The check code is a cyclic redundancy check code, and the cyclic redundancy check code calculates the check value based on the data segment defined by the data length identifier.
[0010] Furthermore, the paperless operation data is decoded and restored from the mixed signal after interference cancellation, including: The transmission protocol unit is parsed from the mixed signal after interference cancellation, and the valid data segment is located based on the data length identifier; The integrity of the valid data segment is verified using the cyclic redundancy check code. When the verification passes, the fragmented data is reassembled according to the serial number identifier to restore the paperless operation data.
[0011] Furthermore, transmission is performed during the low-noise window period at the grid's zero-crossing point, including: Real-time detection of grid voltage waveform to determine zero-crossing point; Using the zero-crossing position as a time reference, a transmission window is formed by extending a preset time to both sides; The analog waveform is sent within the sending window.
[0012] Furthermore, the simulated waveform is coupled to the power distribution line through an impedance matching transformer, and the coupling efficiency is not less than 85%.
[0013] Paperless data transmission system, applied in power distribution line transmission scenarios, including: The noise separation module performs noise separation on the raw electrical signals collected in the power distribution line transmission link to generate a clean signal. The noise separation is achieved through a network model trained based on a deep learning algorithm. The waveform conversion module converts the paperless operation data of the mobile terminal device into an analog waveform suitable for power distribution line transmission, based on the environment of the purification signal. The coupling and transmitting module couples the analog waveform to the power distribution line and transmits it during the low-noise window period of the power grid zero-crossing point. An interference cancellation module, deployed in the terminal equipment of a power distribution line transmission link, reuses the network model to perform noise separation on the received mixed signal in order to eliminate interference; The data restoration module decodes and restores the paperless operation data from the mixed signal after interference elimination.
[0014] Paperless data transmission device, including processor, memory and bus; The memory stores computer program instructions executed by the processor; The processor calls the instructions and data in the memory to implement the paperless data transmission method described above. The bus connects the processor and the memory and is used for transmitting instructions and data interaction.
[0015] The technical solution of this invention can achieve the following technical effects: In this invention, noise separation is achieved through a deep learning network model, directly separating noise features from the original electrical signal to improve the integrity of business data, thereby eliminating the direct impact of interference on data transmission and ensuring channel quality in subsequent processing stages. The paperless operation data of the mobile device is converted into a simulated waveform suitable for transmission over power distribution lines. After being coupled to the power distribution lines, this simulated waveform is transmitted during the low-noise window period at the zero-crossing point of the power grid. This allows data transmission during stable low-noise periods in the power frequency cycle, avoiding instantaneous interference caused by load fluctuations and enhancing transmission reliability.
[0016] To further achieve low-latency transmission, the receiving end directly calls the trained model, eliminating the real-time training process, significantly reducing processing latency, and finally quickly reconstructing paperless operation data based on the clean signal to meet real-time requirements. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 A flowchart of a paperless data transmission method; Figure 2 A flowchart for performing noise separation on the original electrical signal; Figure 3 A flowchart for converting paperless operation data from mobile devices into analog waveforms suitable for power distribution line transmission in an environment based on clean signal; Figure 4 This is a flowchart for decoding and restoring paperless operation data from the mixed signal after interference cancellation. Detailed Implementation
[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.
[0020] Example 1 like Figure 1 As shown, the paperless data transmission method uses power distribution lines for transmission, including: S1: Perform noise separation on the raw electrical signals collected in the power distribution line transmission link to generate cleaned signals. Noise separation is achieved through a network model trained based on deep learning algorithms. S2: Based on the clean signal environment, convert the paperless operation data of mobile devices into analog waveforms suitable for power distribution line transmission; S3: Couple the analog waveform to the power distribution line and transmit it during the low-noise window period of the power grid zero crossing; S4: A multiplexing network model for terminal equipment in power distribution line transmission links, which performs noise separation on the received mixed signals to eliminate interference; S5: Decode and restore the paperless operation data from the mixed signal after interference cancellation.
[0021] In this embodiment, noise separation in step S1 is achieved through a deep learning network model, directly separating noise features from the original electrical signal to improve the integrity of service data, thereby eliminating the direct impact of interference on data transmission and ensuring channel quality in subsequent processing stages. In step S2, the paperless operation data from the mobile device is converted into a simulated waveform suitable for power distribution line transmission. After this simulated waveform is coupled to the power distribution line, it is transmitted in step S3 during the low-noise window period at the zero-crossing point of the power grid. This utilizes the stable low-noise period within the power frequency cycle to transmit data, avoiding instantaneous interference caused by load fluctuations and enhancing transmission reliability.
[0022] To further achieve low-latency transmission, in step S4, the receiving end directly calls the trained model, eliminating the real-time training process and significantly reducing processing latency. Finally, in step S5, the paperless operation data is quickly restored based on the clean signal to meet real-time requirements.
[0023] As a preferred embodiment of the above, such as Figure 2 As shown, the network model is a convolutional recurrent network model, which performs noise separation on the original electrical signal, including: S11: The original electrical signal is input into the convolutional layer to extract the time-frequency domain features of the noise and generate a time-frequency feature tensor. For example, the time-frequency domain features include, but are not limited to, narrowband interference pulse peaks and power frequency harmonic periodic patterns. The time-frequency feature tensor includes, but is not limited to, the spatial distribution information and temporal dynamic information of the noise, providing a structured representation for subsequent separation. The original electrical signal in this step contains service data and noise. As a specific implementation, the convolutional layer scans the signal waveform through a multi-scale filter bank. As a specific implementation, the multi-scale filter bank of the convolutional layer performs a joint time-frequency scan of the original electrical signal, capturing the local features of the narrowband interference pulse peaks and power frequency harmonic periodic patterns, and structurally integrating these features in the three-dimensional space of time-frequency-amplitude to form a time-frequency feature tensor that simultaneously contains the spatial distribution and temporal dynamic evolution of the noise. S12: Input the time-frequency feature tensor into the recursive layer to capture the time dependency of the noise components and generate a noise dynamic feature sequence. As a specific implementation, the recursive layer works by memorizing long-term dependencies based on a gating mechanism. For example, the time dependency includes, but is not limited to, the duration of transient noise caused by load switching and the phase drift of periodic interference. Through this step, the convergence delay problem of traditional adaptive filtering is overcome, and the noise change trajectory is accurately predicted through time series modeling. The noise dynamic feature sequence implies the causal correlation of the noise, ensuring that the separation result conforms to the physical characteristics of power line noise. S13: The noise dynamic feature sequence is decoded by the fully connected layer, and the reconstructed interference waveform is output. As a specific implementation, the fully connected layer decodes through nonlinear transformation, learns the mapping relationship from feature space to time domain waveform, and outputs reconstructed interference that is aligned with the noise component waveform in the original electrical signal. The reconstructed interference retains the time domain form of the actual noise. S14: Directly subtract the interference waveform from the original electrical signal to generate a purified signal.
[0024] In step S14, the subtraction operation is performed directly in the time domain with zero phase offset and no filter group delay, ensuring the integrity of service data.
[0025] As a preferred embodiment of the above, the noise separation performed by the network model on the received mixed signal can adopt the technical solution of steps S11 to S14. In the implementation process, the input to the convolutional layer is the mixed signal, and the interference waveform is finally directly subtracted from the mixed signal. This will not be elaborated here.
[0026] As a preferred embodiment of the above, such as Figure 3 As shown, based on the clean signal environment, the paperless operation data of the mobile device is converted into an analog waveform suitable for power distribution line transmission, including: S21: Encapsulate the paperless operation data into transmission protocol units containing at least a frame header, checksum, and sequence number identifier to generate standard data blocks; through this step, the mobile terminal paperless operation data is divided into data fragments, and the start boundary is located by adding a frame header, the data integrity is verified by adding a checksum, and the sequence number identifier is used to ensure the transmission order. The data is then encapsulated into standard data blocks with the ability to resist transmission errors, providing a structured input basis for subsequent modulation. S22: Perform signal analysis on the purification signal to identify frequency bands with a signal-to-noise ratio higher than the threshold, strong interference frequency bands of frequency converters or inverters, and high-risk noise frequency bands, thereby providing accurate frequency domain basis for dynamic modulation decisions; S23: Use high-order QAM modulation to improve data throughput in frequency bands where the signal-to-noise ratio is higher than the threshold, switch to spread spectrum modulation to enhance noise immunity in strong interference frequency bands of frequency converters or inverters, and convert standard data blocks into modulated waveforms; the modulated waveforms in this step are matched with real-time channel conditions to achieve a dynamic balance between spectral efficiency and reliability. S24: Convert the modulated waveform into an analog baseband waveform synchronized with the power frequency cycle; as a specific implementation method, the digital modulated waveform can be converted into an analog baseband signal by using phase-locked loop technology to make it strictly synchronized with the power frequency cycle of the power grid, generating a physical layer waveform that is aligned with the zero-crossing point of the power grid in the time dimension, and avoiding phase jitter interference caused by load fluctuations. S25: Filter out frequency components corresponding to high-noise frequency bands from the analog baseband waveform and enhance the energy distribution during the low-noise window period of the power grid zero-crossing point, outputting an optimized analog waveform. This step completes the frequency domain shaping of the synchronized analog baseband waveform, ultimately outputting an analog waveform with impedance matching and anti-interference optimization.
[0027] In the above-mentioned optimization method of this application, the modulation strategy is adaptively switched based on the channel analysis results to improve the transmission efficiency in the high-quality frequency band and enhance the anti-noise capability in the interference frequency band, thereby achieving the optimal balance between rate and reliability. Power frequency periodic synchronization technology is used to eliminate power grid phase jitter, generate a time-domain stable analog baseband waveform, avoid load fluctuation interference, and combine frequency domain filtering to remove high-risk noise components and zero-crossing time domain energy focusing to output impedance matching and anti-interference optimization of the final analog waveform, which is directly adapted to the power distribution line transmission environment.
[0028] As a preferred embodiment of the above, the signal analysis execution cycle is 0.5 seconds, and real-time re-analysis is triggered when a start-up or shutdown event of the frequency converter or inverter is detected. In this preferred scheme, the channel analysis results are dynamically updated through a fixed period and an event triggering mechanism to ensure that the modulation strategy adapts to sudden interference changes in new energy power plants in real time.
[0029] As a preferred embodiment of the above, the transmission protocol unit further includes a data length identifier; the check code adopts a cyclic redundancy check code, and the cyclic redundancy check code calculates the check value based on the data segment defined by the data length identifier.
[0030] In this preferred scheme, the data length identifier clearly defines the byte range of valid data, and the cyclic redundancy check code accurately covers the service content within this length. The two work together to prevent boundary confusion caused by frame structure parsing errors and resist bit errors caused by power line noise through a strong check mechanism, which significantly improves the integrity and reliability of the transmission protocol unit under poor channel conditions.
[0031] As a preferred embodiment of the above, such as Figure 4 As shown, the data for paperless operations is decoded and restored from the mixed signal after interference cancellation, including: S51: Extract transmission protocol units from the mixed signal after interference cancellation and locate valid data segments based on data length identifiers; S52: Cyclic redundancy check (CRC) is used to verify the integrity of valid data segments; S53: When the verification passes, reassemble the fragmented data according to the serial number identifier to restore the paperless operation data.
[0032] In this preferred solution, precise positioning through data length identifiers can avoid frame boundary misalignment caused by noise residue. The receiver reuses the data length identifiers and cyclic redundancy check code rules defined by the transmitter to achieve zero-configuration self-adaptation, enabling lossless restoration of business data in noisy environments.
[0033] As a preferred embodiment of the above, transmission during the low-noise window period of the power grid zero-crossing point includes: S31: Real-time detection of grid voltage waveform to determine zero-crossing position; S32: Using the zero-crossing position as the time reference, a transmission window is formed by extending a preset duration to both sides; S33: Send an analog waveform within the sending window.
[0034] In this preferred scheme, by accurately locking the zero-crossing point of the power grid and expanding the symmetrical transmission window, the instantaneous interference caused by load fluctuations is avoided, ensuring that the analog waveform is transmitted during the most stable low-noise period within the power frequency cycle.
[0035] In practice, preferably, the analog waveform is coupled to the power distribution line through an impedance matching transformer, with a coupling efficiency of not less than 85%. The impedance matching transformer is used to achieve efficient energy transfer, minimize signal reflection loss, and ensure the energy integrity when the analog waveform is injected into the power distribution line.
[0036] Example 2 Paperless data transmission system, applied in power distribution line transmission scenarios, including: The noise separation module performs noise separation on the raw electrical signals collected in the power distribution line transmission link to generate clean signals. The noise separation is achieved through a network model trained based on deep learning algorithms. The waveform conversion module, based on the clean signal environment, converts the paperless operation data of the mobile device into an analog waveform suitable for power distribution line transmission; The coupling and transmitting module couples the analog waveform to the power distribution line and transmits it during the low-noise window period of the power grid's zero-crossing point; Interference cancellation module, deployed in the terminal equipment of power distribution line transmission link, uses a multiplexed network model to perform noise separation on the received mixed signal in order to eliminate interference; The data restoration module decodes and restores the paperless operation data from the mixed signal after interference cancellation.
[0037] Example 3 Paperless data transmission device, including processor, memory and bus; Memory stores computer program instructions executed by the processor; The processor calls instructions and data from memory to implement the paperless data transmission method as described in Embodiment 1; The bus connects the processor and memory, and is used for transmitting instructions and exchanging data.
[0038] The technical effects achieved in Embodiments 2 and 3 are as described in Embodiment 1 above, and will not be repeated here.
[0039] Those skilled in the art should understand that this invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to this invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.
Claims
1. A paperless data transmission method, characterized in that, Power transmission via power distribution lines includes: Noise separation is performed on the raw electrical signals collected in the power distribution line transmission link to generate a cleaned signal. The noise separation is achieved through a network model trained based on a deep learning algorithm. Based on the environment of the purification signal, the paperless operation data of the mobile terminal device is converted into a simulated waveform suitable for power distribution line transmission; The simulated waveform is coupled to the power distribution line and transmitted during the low-noise window period of the power grid zero-crossing point; The terminal equipment of the power distribution line transmission link reuses the network model to perform noise separation on the received mixed signal in order to eliminate interference; The paperless operation data is restored by decoding the mixed signal after interference elimination.
2. The paperless data transmission method according to claim 1, characterized in that, The network model is a convolutional recurrent network model, and the noise separation process for the original electrical signal includes: The original electrical signal is input into the convolutional layer to extract the time-frequency domain features of the noise and generate a time-frequency feature tensor. The time-frequency feature tensor is input into the recursive layer to capture the time dependency of the noise components and generate a noise dynamic feature sequence. The noise dynamic feature sequence is decoded by a fully connected layer, and the reconstructed interference waveform is output. The purified signal is generated by directly subtracting the interference waveform from the original electrical signal.
3. The paperless data transmission method according to claim 1, characterized in that, Based on the environment of the purification signal, the paperless operation data of the mobile device is converted into an analog waveform suitable for power distribution line transmission, including: The paperless operation data is fragmented and encapsulated into transmission protocol units containing at least a frame header, a checksum, and a sequence number identifier to generate standard data blocks; The purification signal is analyzed to identify frequency bands with a signal-to-noise ratio higher than the threshold, strong interference frequency bands of frequency converters or inverters, and high-risk noise frequency bands. High-order QAM modulation is used in the frequency band where the signal-to-noise ratio is higher than the threshold, and spread spectrum modulation is switched in the strong interference frequency band of the frequency converter or inverter to convert the standard data block into a modulated waveform. The modulated waveform is converted into an analog baseband waveform synchronized with the power frequency cycle; The frequency components corresponding to the high-risk noise bands are filtered out from the simulated baseband waveform, and the energy distribution during the low-noise window period of the power grid zero crossing is enhanced to output the optimized simulated waveform.
4. The paperless data transmission method according to claim 3, characterized in that, The signal analysis has an execution cycle of 0.5-0.8 seconds, and real-time reanalysis is triggered when a start-up or shutdown event of the frequency converter or inverter is detected.
5. The paperless data transmission method according to claim 3, characterized in that, The transmission protocol unit also includes a data length identifier; The check code is a cyclic redundancy check code, and the cyclic redundancy check code calculates the check value based on the data segment defined by the data length identifier.
6. The paperless data transmission method according to claim 5, characterized in that, Decoding and reconstructing paperless operation data from the mixed signal after interference cancellation includes: The transmission protocol unit is parsed from the mixed signal after interference cancellation, and the valid data segment is located based on the data length identifier; The integrity of the valid data segment is verified using the cyclic redundancy check code. When the verification passes, the fragmented data is reassembled according to the serial number identifier to restore the paperless operation data.
7. The paperless data transmission method according to claim 1, characterized in that, Transmitted during the low-noise window period of the power grid zero-crossing point, including: Real-time detection of grid voltage waveform to determine zero-crossing point; Using the zero-crossing position as a time reference, a transmission window is formed by extending a preset time to both sides; The analog waveform is sent within the sending window.
8. The paperless data transmission method according to claim 1 or 7, characterized in that, The simulated waveform is coupled to the power distribution line through an impedance matching transformer, and the coupling efficiency is not less than 85%.
9. A paperless data transmission system, characterized in that, Applications include power distribution line transmission scenarios, including: The noise separation module performs noise separation on the raw electrical signals collected in the power distribution line transmission link to generate a clean signal. The noise separation is achieved through a network model trained based on a deep learning algorithm. The waveform conversion module converts the paperless operation data of the mobile terminal device into an analog waveform suitable for power distribution line transmission, based on the environment of the purification signal. The coupling and transmitting module couples the analog waveform to the power distribution line and transmits it during the low-noise window period of the power grid zero-crossing point. An interference cancellation module, deployed in the terminal equipment of a power distribution line transmission link, reuses the network model to perform noise separation on the received mixed signal in order to eliminate interference; The data restoration module decodes and restores the paperless operation data from the mixed signal after interference elimination.
10. A paperless data transmission device, characterized in that, Includes processor, memory, and bus; The memory stores computer program instructions executed by the processor; The processor calls the instructions and data in the memory to implement the paperless operation data transmission method as described in any one of claims 1-8; The bus connects the processor and the memory and is used for transmitting instructions and data interaction.